A Tunable Base Station Cooperation Scheme for Poisson Cellular Networks

TL;DR

This paper introduces a flexible, location-dependent base station cooperation scheme for Poisson cellular networks, optimizing SIR and spectral efficiency by tuning cooperation levels across different cell regions.

Contribution

It presents a novel tunable cooperation scheme based on stochastic geometry, with analytical expressions for SIR distribution and spectral efficiency, enhancing network performance.

Findings

Moderate cooperation improves SIR performance.

The scheme maintains spectral efficiency while enhancing SIR.

Analytical models accurately predict performance gains.

Abstract

We propose a tunable location-dependent base station (BS) cooperation scheme by partitioning the plane into three regions: the cell centers, cell edges and cell corners. The area fraction of each region is tuned by the cooperation level $\gamma$ ranging from 0 to 1. Depending on the region a user resides in, he/she receives no cooperation, two-BS cooperation or three-BS cooperation. Here, we use a Poisson point process (PPP) to model BS locations and study a non-coherent joint transmission scheme, $\textit{i.e.}$, selected BSs jointly serve one user in the absence of channel state information (CSI). For the proposed scheme, we examine its performance as a function of the cooperation level using tools from stochastic geometry. We derive an analytical expression for the signal-to-interference ratio (SIR) distribution and its approximation based on the asymptotic SIR gain, along with the characterization of the normalized spectral efficiency per BS. Our result suggests that the proposed scheme with a moderate cooperation level can improve the SIR performance while maintaining the normalized spectral efficiency.